Automatic Normalization of Word Variations in Code-Mixed Social Media Text
This addresses the challenge of processing noisy, multilingual social media data for NLP applications, but it is incremental as it builds on existing methods for normalization.
The paper tackled the problem of word variations in code-mixed social media text by using unsupervised distributed representations to normalize spelling variations, resulting in improved performance in part-of-speech tagging and sentiment analysis tasks.
Social media platforms such as Twitter and Facebook are becoming popular in multilingual societies. This trend induces portmanteau of South Asian languages with English. The blend of multiple languages as code-mixed data has recently become popular in research communities for various NLP tasks. Code-mixed data consist of anomalies such as grammatical errors and spelling variations. In this paper, we leverage the contextual property of words where the different spelling variation of words share similar context in a large noisy social media text. We capture different variations of words belonging to same context in an unsupervised manner using distributed representations of words. Our experiments reveal that preprocessing of the code-mixed dataset based on our approach improves the performance in state-of-the-art part-of-speech tagging (POS-tagging) and sentiment analysis tasks.